Foldit Wiki
Advertisement
CASP 16 L1000 ligand queue

Puzzle 2452, showing the ligand queue tool in action.

In the world of proteins, CASP is an acronym for Critical Assessment of Structure Prediction. CASP is a bi-annual worldwide experiment/competition involving protein structure prediction, held since 1994.

The official CASP site is www.predictioncenter.org, and has the results of previous competitions and also information on the current CASP.

The wikipedia article CASP has some additional background, and describes the competitive aspects in more detail than the predictioncenter site.

Foldit players participated in several CASP competitions, but recent CASPs have involved large proteins that aren't a good match for Foldit's limitations.

CASP 16 (2024)

After nearly 10 years, Foldit returns to CASP, with Puzzle 2452: CASP16 Ligand Puzzle L1000.

Puzzle 2452 presents 17 predefined ligands and a 247-residue protein. Unlike recent small molecule design puzzles, the ligand can't be changed, but the protein is completely unlocked, allowing use of Remix and other tools that aren't normally useful in small molecule puzzles.

The goal in puzzle 2542 is to find the best docking position for each of the 17 ligands. The protein has been solved multiple times. The expectation is that a successful solution won't change the shape of the protein very much.

CASP 11 (2014)

Foldit players submitted results for many targets in CASP11. See CASP 11 Rankings here on the wiki for the results.

CASP Roll

The official CASP Roll page

As the name implies, CASP roll was a rolling experiment, which lasted for an entire year.

  • Table of all puzzles and natives. (German foldit wiki)
  • The result of puzzles previous to casp10 was published.

All Results

CASP10

See CASP10 for the official results.

Timeline:

  • The  prediction targets are released
  • the last prediction targets have been released in July 17, 2012
  • prediction season end July 31, 2012
  • Refinement experiment end in August 17, 2012
  • Abstracts describing the methods tested in CASP10 were collected in September 2012
  • See the table on the German Foldit wiki for results of all CASP10 puzzles.
  • See some statitics from spmm.
  • See some results of Foldit players
  • blog entries: CASP10 update (2012/09)
  • Group performance on all models #101 Foldit, #124 AD, #127 Contenders, #140 VC
  • Group performance on first model #116 Foldit, #120 VC, #123 AD, #126 Contenders
  • WeFold results : WeFold  wfFUIK, (Foldit participating in the WeFold CASP 10 project )

Mini-CASP

In early August of 2009, the foldit developers decided to give the players starting models outputted from Rosetta@home, by obtaining sequences for proteins that were currently unsolved, but would be released to the Protein Data Bank soon after. Since no one knew which structure predictions would be correct, they tried to give the players as diverse a set as possible.

The intent was practice for everyone for the upcoming CASP9, and to help the developers figure out what the best Rosetta models would be to use as starting points for Foldit. (taken from Dr. Baker's Mini CASP announcement http://fold.it/portal/node/986632)

CASP9

See CASP 9 for results.

CASP8

Foldit was represented for the first time at the eighth competition (CASP8) which took place during 2008, but in that competition FoldIt results were part of the BakerLab CASP team. In preparation the players had folded old CASP7 proteins and compared the results from the developers with the published structures.

The first open beta of FoldIt was released a few days after the start of CASP8[1]. The players and developers then spent a month training and developing the program further before trying their hand at their first CASP8 puzzle[2]. As a result, they were only involved in relatively few competition targets. Despite these handicaps the FoldIt team did well with some top places and even a victory! Overall, the Foldit players were as good as the experts with all their tools and better than any automated machine submissions[3].

Literature

Building the Scientific - bild der wissenschaft edition: 4 / 2010, page 18 - Perfectly shaped - Article on Protein Structure Prediction and CASP

Sources


Advertisement